The Estonian E-Voting Laws Discourse:
for Central and Eastern Europe
The Republic of Estonia has been, and still is, widely credited to be a pioneer in e-governance and
especially e-democracy, with headlines such as “Estonia: 10 Years from Communism to Advanced e-
It had frequently been expected, too, that Estonia would be the leading country for e-
voting, introducing it already for the national elections this year.
However, in the very last changes of
the respective laws, the Estonian Parliament voted for e-voting, not for the immediate future, but only
with a delay of implementation until the year 2005. Still, the Estonian is the first case world-wide of a
country that has actually passed overall e-voting laws.
This primacy, by virtue of some variant of ‘the normative power of the factual’, therefore sets the scene
for all e-voting laws considered anywhere – but especially so for the countries of Central and Eastern
Europe (CEE), if we believe at all in any form of regional characteristics and therefore similarity and
comparability. The Estonian e-voting laws, but also the discourse through which they emerged,
therefore serve as a paradigmatic benchmark – broadly understood here as a benchmark both in a
positive as well as possibly in a negative sense as well – for all e-voting law discourses to occur in all
the other CEE countries.
This paper therefore sets out to, first, follow the process – interesting as such, it has only been
documented and analyzed once before (Drechsler and Madise 2002), and the current piece presents a
University of Tartu, Estonia.
In the article at http://www.e-
Statements, state of legislation, and web-links of this paper are generally valid as of 15 February 2003.
I am grateful to Rainer Kattel and Ülle Madise for comments on this paper.
See, e.g., http://www.time.com/time/interactive/stories/society/e_politics.html;
“You can’t stop progress though and it looks
as though Estonia will be the world’s first nation to provide e-voting at its next General Election in
very short but also updated account of that story.
The question of discourse will then be addressed,
and an interpretation offered. As the benchmarking will indeed be seen to entail an inverted, i.e.
negative, quality as well – the dog that did not bark during the night –, a final segment will also briefly
discuss these aspects.
2. E-voting in Estonia: a narrative
The plan to introduce e-voting in Estonia was first publicly announced by the Minister of Justice, Märt
Rask, a member of Reformierakond (“Reform Party”, eng.reform.ee
), the neo-liberal (indeed, market-
radical) ‘transition winners’’ party, at the beginning of 2001. Given the general fashion of e-related
matters, which is particularly strong in Estonia, and swift developments in such fields as e-banking (see
), paperless government (see
), and broadcasting of
parliamentary sessions (see www.riigikogu.ee/news.html
), this was a likely step to take.
Estonian leadership in e-related fields was and is also seen as a key part of ‘branding Estonia’ and
overall of making Estonia better known globally.
The idea of e-voting was thus strongly promoted by Prime Minister Laar, who in the Parliamentary
question time of 17 January 2001 proposed the idea of testing e-voting during the same year and to
decide then whether to introduce it already for the 2002 local elections. (See
To get an overview of the possible methods and risks of remote Internet voting, the Ministry of Justice
ordered an analysis from two scholars in the field. (Lipmaa and Mürk 2001) The report by the Internet
Policy Institute published in the USA at the same time
was also used as a basis of study. The
commissioned analysis recommended to prepare some experiments or pilot-projects first and not to
introduce e-voting before 2007, because an earlier date would be technically, and therefore also
socially, too risky. (1, 28-30) In the fall of 2001, another analysis was ordered from a mathematician
by the Estonian Ministry of Transport and Communication, which was to focus especially on technical
questions and costs. In this analysis, concrete recommendations concerning the voting process were
given and a provisional budget of e-elections was drawn up. (IT Meedia 2001)
I would like to thank the co-author of that essay, Ülle Madise, without whom the current paper
would not exist, for her very kind permission to use materials of the earlier piece, unmarked, in the
current one. Drechsler and Madise 2002 also contains extensive references and technical as well as
See generally, e.g., the Estonian Agenda 21, www.agenda21.ee/english/EA21/3_4.html.
The Estonian e-voting law discourse
Taking into account the purpose to introduce public remote Internet voting and some of the
recommendations given by the experts – but not the one by its own experts as regarded postponement
until 2007 –, e-voting provisions were drafted by the Ministry of Justice and sent to the parliament.
There, they were not discussed generally, but as part of four different new election laws: The Local
Communities Election Act, the Referendum Act, the European Parliament Election Act, and the
Riigikogu Election Act. The discussions in the Riigikogu as far as the e-voting feature was concerned
were more or less seamless as well and not really closely connected with what Act it actually was.
However, since local elections were scheduled for 2002, it was this Act that drew more attention than
the other, followed by the Referendum Act because of its implications for European Union accession.
When discussing specific features (which in the end were the same for all four Acts), I will therefore
refer below to the development of the provisions of the Local Communities Election Act.
As was to be expected, already in the very first stage of developing the e-voting idea, the old and new
government coalition parties – until January 2002, Right-Libertarian-Moderate, from then Libertarian-
Populist – were principally in favour of e-voting, the opposition parties Rahvaliit and Ühendatud
Rahvapartei factions against. In order to understand this, it is important to briefly sketch out the
Estonian party structure. The first government coalition mentioned here (‘Laar II’ – the government
headed, for the second time, by Mart Laar) included, in addition to the already-mentioned ‘libertarian’
– the ‘right’ Isamaaliit (“Pro Patria Union”, www.isamaaliit.ee/isamaa2/index_eng.html
; cf. also
Laar 2002), a generally nationalist but for the most part also market-radical party that in a
slightly different composition formed the government right after the regaining of
– the ‘moderate’ Mõõdukad (“Moderates”, www.moodukad.ee/
), who by their self-definition are
Social Democrats but by ‘Western’ standards quite to the right of that field.
The second government coalition (‘Kallas’ – headed by Siim Kallas) includes Reformierakond and
– the ‘populist’ Keskerakond (“Centre Party”, www.keskerakond.ee/
), the main ‘transition
losers’’ party, with a semi-charismatic leader, Edgar Savisaar, currently the Mayor of Tallinn,
but without a genuine post-Socialist ideology.
The two opposition parties mentioned are
– Eestimaa Rahvaliit (“Estonian Peoples Union”, www.erl.ee
), a party similar to Keskerakond
but with a strong and explicit rural orientation; and
– Eestimaa Ühendatud Rahvapartei (“Estonian United Peoples Party”, www.eurp.ee/eng/
most clearly post-Socialist party with a special appeal for that part of the Russian-speaking
population of Estonia that is actually allowed to vote.
The initial draft of Local Communities Election Act can be found at
: Täiskogul menetletud eelnõu nr 747. Menetlusetapid. Algtekst.
The governing coalition Kallas does not command a majority in the Riigikogu – rather, only 47 of the
101 votes. The missing votes are usually delivered by the Rahvaliit (which is also the party of the
President), so that party cannot be ignored.
E-voting provisions were always supported in plenary session.
It should be noticed that pilot projects
were perhaps occasionally considered, as in the beginning by Laar himself, but they were never
seriously put on the agenda. In some sense, the entire draft and then law would be its own pilot project
– not a rare modus operandi in Estonia. In the end, as a form of compromise, in all laws or drafts,
was explicitly stated that e-voting should not be applied before the year 2005 (§ 74 (5) in the Local
Communities Election Act). This was apparently in deference to the Rahvaliit faction – as was
mentioned, the government commands at best a minority of 47 out of 101 votes, and their Rahvaliit
votes are therefore usually important (although not in this special case), which is why their opinion is
taken into consideration.
3. The discourse
The discourse to be analyzed already starts on the Ministerial level. According to § 60 of the Estonian
Constitution, “Members of the Riigikogu shall be elected in free elections on the principle of
proportionality. Elections shall be general, uniform and direct. Voting shall be secret.” Since the
original drafting of the Constitution of 1992, these principles have not been the subject of juridical
discussion, so they are ill-defined. (See Annus 2001, 64-70) As to whether e-voting would influence
these principles, the Minister and Ministry based themselves on two basic decisions:
According to the descriptions given above, this means that the current governing coalition
consists of ‘transition winners’ and ‘transition losers’. However, in Estonia this is not necessarily a
contradiction, because Reformierakond indeed does promulgate an ideology appropriate for its
clientele, but Keskerakond does not; rather, it has hardly any ideology at all – it is, therefore, a classic
populist party. They are therefore not unlikely coalition partners at all. – Votes are scheduled for
March 2003, i.e. they fall between the deadline for this paper and the Bucharest conference, so there
might easily be a new coalition in place once this paper is discussed.
See the debate and voting results according to the minutes as cited below (FN 10). About the
voting process, see the Riigikogu Internal Rules Act, www.riigikogu.ee/legislation.html
The Local Communities Election Act was adopted by Parliament on 27 March 2002 and
entered into force on 6 May 2002. RT I 2002, 36, 220. The Referendum Act was adopted on 13 March
2002 and entered into force on 6 April 2002. RT I 2002, 30, 176. The Riigikogu Election Act was
adopted on 12 June 2002 and entered into force on 18 July 2002. RT I 2002, 57, 355. The European
Parliament Election Act was adopted by Parliament on 18 December 2002 and entered into force on 23
January 2003. RT I 2003, 4, 22.
The Estonian e-voting law discourse
1. To use a teleological approach to Constitutional interpretation, i.e. to say that Constitutional
problems should be understood through the problems the given principles were meant to
solve. As an example in the current case of e-voting, the principle of secrecy (raised most
strongly in Parliament later on) was said to protect an individual from any pressure or
influence against her or his free expression of the political preference – i.e., that it is a means,
not an end. This includes the threat that the state or a public official can check who voted for
whom. But it was said that, if privacy is guaranteed in the polling station and if all those who
have voted via the Internet have the right (which was proposed) to go to the polling station on
election day and replace their electronically recorded, transferred and counted vote by a new
paper-ballot (see § 55 of the initial draft of the Local Communities Election law), then the aim
of the principle of secrecy, the end, is actually achieved.
2. To start from the assumption that the State must ‘trust the people’ and not interfere if at all
possible in any of their decisions. The Reformierakond ideology informs this approach. As an
example in our context, the problem that e-voting would facilitate some families, friends or
colleagues voting together, i.e. practice collective voting, as well as the buying and selling of
votes, was said to hinge on the question of whether the State would have to protect an
individual only from other individuals or also from her- or himself. It was not seen that
collective voting could be a problem for the state as well, and not only for the individual.
Parliamentary debate on e-voting was long and lively. In the plenary session, e-voting was discussed
within all readings of all four drafts.
We can draw up the following tables of discussion points of
problems of e-voting:
1. Equality of citizens in political life – “unfair” towards non-connected citizens / digital gap
2. Detriment to democracy (going to the polling station would be a valuable action by itself)
3. Unconstitutionality of e-voting (secrecy, generality, and uniformity)
4. Privacy and secrecy of voting not guaranteed
5. Security of electronic voting systems not sure
6. Proneness to fraud
7. Negative or absent experiences in other countries
8. The weakness of technical preparations
9. The problem of hackers
See the minutes at www.riigikogu.ee/ems/index.html. The draft of the Local Communities
Election was discussed on 14 June 2001, 23 January, 27 February, and 27 March 2002; the draft of the
Riigikogu Election Act on 14 June 2001, 30 January 2002, 27 March, 15 and 22 May 2002; the draft of
the Referendum Act on 19 September 2001, 30 January and 13. March 2002; the draft of the European
Parliament Election Act on 23 January 2002.
It may generally be noticed that a large majority of Members shared the Ministry’s attitude towards
teleological interpretation of the Constitution, as well as the assumption that
1. e-voting increases voter turnout; and that this
2. automatically has a positive effect on ‘Democracy’.
There was hardly any accompanying discussion of e-voting in media or society (with the exception of a
few newspaper articles and simple and emotional anonymous comments to them in online-newspapers
likewise, neither were there any significant public comments by social scientists or
lawyers. In January 2001, the editorial of the business daily Äripäev had been devoted to the idea of the
Minister of Justice to introduce e-voting in Estonia. (“Miks oodata aastani 2003” 2001) The editor
asked why Estonia should wait until 2003; rather, Internet voting should be introduced already for the
local elections of 2002. This had then been discussed. (See Äripäev Online, 5 January 2001,
One can safely say that the e-voting initiative came from political elite, and that it was and is largely
detached from ‘the people’ whose participation it is supposed to increase. One could certainly
diagnose for Estonia an attitude towards the right to vote, and democratic decision-making in general,
that one might describe variously as pragmatic, relaxed, detached, or cynical. Anecdotally, as regards
e.g. possibilities of fraud, one could often hear people saying that, if they trusted the net with their
banking, why should they not in a so much less important field as political elections?
Still, while Estonia could have easily been the world leader in e-voting by introducing this as a regular
feature already for the local elections of 2002, probably genuine worries that technical problems would
not be solved by the Fall of that year, as well as the scepticism of individual members of parties
generally in favour of e-voting, all of them reasonable and appropriate, were among the reasons that
prevented such an outcome. Nonetheless, the resistance of the rural opposition party, which – likewise
reasonably and appropriately – feared that such a feature would increase the vote of its competitor
parties, and which therefore would have very rightly and properly fought against it in Parliament, at
least contributed significantly to the postponement of actual e-voting in Estonia until 2005.
However, many of the to-be-expected points of discourse (Will 2002 and Buchstein and Neymanns
2002 provide very good surveys) were hardly considered, and are missing from the Estonian discourse.
Without claiming completeness, I would want to single out the following five points, two general and
See, e.g., www.postimees.ee; www.delfi.ee. As all comments are anonymous, their level is
indeed exceedingly low, and they often do not connect with the subject at hand.
The Estonian e-voting law discourse
1. Are the effects of e-voting really beneficial for Democracy?
2. Will e-voting increase voter turnout?
3. How high are really the costs?
4. Are there possibly adverse effects of the e-voting provisions for joining the European Union?
5. Are there dangers of a law suit on the basis of the European Human Rights Convention?
4. and 5. came as a great surprise to many Estonian experts when mentioned; as regards 3., out of all
people, the “internet guru” of the Laar administration, Linnar Viik, cautioned that, compared to a
traditional one, “It'll cost ten times as much to have an e-election,”
but it did hardly enter the
More interesting for this paper, however, are the more fundamental questions of 1. and also of 2.,
because they are so often badly considered. Regarding these, results are not really ‘in’ yet, but
naturally, in such a key matter as Democracy, if there is reason for a cautionary approach, this should
be mentioned very clearly, and also discussed and taken into consideration before laws are passed.
Estonia is noticeable for its strong proclivities of anything e-related among its politico-economic elite,
as well as for an extremely low level of resistance against, and indeed discourse about, any
‘progressive’ developments that might have unwanted side-effects (biotechnology is another example;
see Weber 2001), which is perhaps why these matters were comparatively unaddressed. However,
probably not many CEE countries are doing much better in this respect. Thus, the following points
might serve as a general reminder:
4.1. Voter turnout
We have no good reason to think that e-voting will necessarily increase voter turnout. Rather, it seems
that those people who will vote on-line are highly e-literate people who are politically interested
already. (See Kersting and Baldersheim forthcoming) Darin Barney has noticed, correctly, that “recent
research indicates that network technologies tend to reinforce existing patterns of democratic behaviour
rather than mobilizing new actors and practices.” (2001, 264) But even if it were otherwise, one might
also consider theories such as the ‘Crispin Curve’ argument that overly high voter turnout is a sign of
problems, not of a healthy Democracy (1948, 160-165).
Laar’s continuous touting of e-voting as a possibility to increase voter turnout and (partially therefore)
develop democracy (see Laar 2002, 244-246 et passim) is therefore without any rational basis.
4.2. Digital gap
That the Digital Divide or Gap is a real threat that will in all likelihood widen various already-existing
gaps in society is, I think, clear for anyone who has studied the subject, and it has certainly been
Quoted in http://www.thefeature.com/index.jsp?url=article.jsp?pageid=12832; on Viik, see
demonstrated for Estonia. (Kalkun and Kalvet 2002) And if Democracy is about representing people,
then Rahvaliit is right: The studies we have indicate that internet voting can substantially change the
result. (Cf., e.g., Tolbert and McNeal 2001 for the influence of Internet access (without e-voting) on
voter turnout) One of the most recent thorough studies we have, of a German county commissioner
election with model e-voting, shows that the result via e-voting would have brought another candidate
to power (the more left one, incidentally). (See Meuren 2001) It is somehow difficult to reconcile this
with the basic principles of participatory democracy.
4.3. e-matters and Democracy
But there is also a very general problem, which I will address only briefly, and by way of some quotes,
here (see Drechsler 2002 for a more extensive argument). Hubert Dreyfus’ excellent critique of ‘virtual
community’ and ‘electronic republic’ advocates, who suffer from a deep deficiency of thought on what
a polis is about and what are pure incidentals or mechanisms (2001, 103-106), should be mandatory
reading for all dealing with e-governance. As he says, “The Athenian agora is precisely the opposite of
the public sphere, where anonymous electronic kibitzers from all over the world, who risk nothing,
come together to announce and defend their opinions. As an extension to the deracinated public
sphere, the electronic agora is a grave danger to real political community. ... it is ... a nowhere place for
anonymous nowhere people.” (104)
On virtual communities, Darin Barney makes the similar general point: “Though they might feel like it,
the fact remains that computer networks are not real places, and while their virtuality might present
certain benefits for community formation, these same attributes compromise the rootedness of those
communities once they are established.” (2001, 214) “The network digital computer is often presented
to contemporary individuals as the final technology of their ultimate self-creation ... in so far as they
reduce the world – human beings included – to a standing-reserve of bits, networks culminate the
distinctly modern technological conditions described by Martin Heidegger: a condition characterized
by rootlessness, calculation, and the denial of mystery.” (195) Barney cites a 1998 study from Calgary
in which “it was found that membership in network associations had ‘corrosive effects’ on civility:
‘Respondents who were most engaged online tended to be relatively disengaged with (and distrusting
of) the “real” community. It appears that these online associations could be damaging to civil society’”.
It is well known that cyberspace, information and communication technologies (ICT), the internet, the
web, network technology, whatever you call it, makes our lives better, easier, and safer; flattens
hierarchies and thus makes people more independent; fosters democracy; improves social capital and
the sense of community; allows for greater freedom for the individual person because of the possibility
of re-defining oneself again and again, and so on. This is the basis of the desirability of e-voting as
The Estonian e-voting law discourse
well as e-governance; unfortunately, as basic assumption of an automatism, it is also exactly as wrong
as it is well-known. Yet, it can hardly be doubted that, technology-driven as our time is, this is the
‘train into the future’. What the Estonian case shows, and why this is an excellent paradigmatic
benchmark for Central and Eastern Europe, is that one should consider the problems of e-voting
thoroughly before passing respective laws; and the discourse analysis has shown that one can simply
not rely on the assumption that a nice conversation among all stakeholders will happen. The
detachment of the discourse from scientific approaches and study results in just such a science- and
progress-charged field is particularly curious. All this presents a challenge precisely to social scientists
in the area to push for a higher and therefore more responsible level of discourse.
On the most basic level, “When societal consideration of a new technology is limited to identifying
technical problems and technical solutions, the general condition in which technology holds sway is
reinforced rather than challenged. This, by and large, has been the case with network technology.”
(Barney 2001, 233) Worse, it has become part of the general paradigm of today, and even modest
critics of the net easily appear as luddites. The most appropriate counter for this is to take a step back
and look at the issue from the perspective of what the human person can and should be, and then
consider what network technology generally, and e-voting specifically, does.
(Note: Pure web-based information is not reprinted here.)
Annus, Taavi (2001). Riigiõigus. Tallinn: Juura.
Buchstein, Hubertus and Harald Neymanns (Eds.) (2002). Online-Wahlen. Opladen: Leske + Budrich.
Crispin, Edmund (1948). Buried for Pleasure. London: Victor Gollancz.
Drechsler, W. (2002). Review of Dreyfus 2001 and Barney 2001. Philosophy in Review / Comptes
Rendus Philosophiques 22 (2): 87-90.
Drechsler, W. and Ü. Madise (2002). “e-voting in Estonia.” Trames 6 (3): 234-244. Also forthcoming
in Kersting and Baldersheim.
Dreyfus, Hubert L. (2001). On the Internet. London – New York: Routledge.
IT Meedia (Tanel Tammet and Hannes Krosing) (2001). “E-valimised Eesti Vabariigis: võimaluste
analüüs.” Also at http://ats.riik.ee/amphora/home/projektid/e-
Kalkun, Mari and Tarmo Kalvet (Eds.) (2002). Digital Divide In Estonia and How to Bridge It. Tallinn:
Emor and PRAXIS Centre for Policy Studies. E-book, downloadable from
Kersting, Norbert and Harald Baldersheim (Eds.) (forthcoming). Internet Voting. Present Forms and
Future Perspectives. Basingstoke: Palgrave Macmillan.
Laar, Mart (2002). Das estnische Wirtschaftswunder. [Tallinn:] Konrad-Adenauer-Stiftung.
Lipmaa, H. and O. Mürk (2001). “E-valimiste realiseerimisvõimaluste analüüs.” Analysis ordered by
the Estonian Ministry of Justice. Also at www.just.ee/oldjust/JM/lipmaamyrk.pdf
Meuren, D. (2001). “Testfall Landratswahl.” Spiegel Online (17 September), at
“Miks oodata aastani 2003.” Editorial, Äripäev (8 Januar 2001); also at
Tolbert, C. and R. McNeal (2001). “Does the Internet increase Voter Participation in Elections?” Paper
presented to the Annual Meeting of the American Political Science Association, San
Weber, A. (2001). “baltikummer. Das verkaufte Volk”, SZ Magazin (3 November); also at
Will, Martin (2002). Internetwahlen. Verfassungsrechtliche Möglichkeiten und Grenzen. Stuttgart etc.: